Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press A genetic interaction analysis identifies cancer drivers that modify EGFR dependency Sida Liao, Teresa Davoli, Yumei Leng, Mamie Z. Li, Qikai Xu, and Stephen J. Elledge Division of Genetics, Department of Medicine, Brigham and Women’s Hospital, Department of Genetics, Program in Virology, Howard Hughes Medical Institute, Harvard University Medical School, Boston, Massachusetts 02115, USA A large number of cancer drivers have been identified through tumor sequencing efforts, but how they interact and the degree to which they can substitute for each other have not been systematically explored. To comprehensively investigate how cancer drivers genetically interact, we searched for modifiers of epidermal growth factor receptor (EGFR) dependency by performing CRISPR, shRNA, and expression screens in a non-small cell lung cancer (NSCLC) model. We elucidated a broad spectrum of tumor suppressor genes (TSGs) and oncogenes (OGs) that can genetically modify proliferation and survival of cancer cells when EGFR signaling is altered. These include genes already known to mediate EGFR inhibitor resistance as well as many TSGs not previously connected to EGFR and whose biological functions in tumorigenesis are not well understood. We show that mutation of PBRM1, a subunit of the SWI/SNF complex, attenuates the effects of EGFR inhibition in part by sustaining AKT signaling. We also show that mutation of Capicua (CIC), a transcriptional repressor, suppresses the effects of EGFR inhibition by partially restoring the EGFR-promoted gene expression program, including the sustained expression of Ets transcription factors such as ETV1. Together, our data provide strong support for the hypothesis that many cancer drivers can substitute for each other in certain contexts and broaden our understanding of EGFR regulation. [Keywords: cancer drivers; EGFR; genetic interaction] Supplemental material is available for this article. Received October 11, 2016; revised version accepted January 3, 2017. Cancer is driven by a number of distinct genetic alter- 2011), it is currently unclear how many cancer genes oper- ations, including gain or loss of chromosomes and chro- ate to achieve these conditions. From a theoretical per- mosomal segments, translocations, and point mutations spective, it makes sense that many cancer drivers may that result in inactivation of tumor suppressor genes perform similar functions and be partially interchange- (TSGs) or activation of oncogenes (OGs). Attempts to able during tumor evolution. They may act either to ge- identify these cancer drivers based on patterns of muta- netically modify a shared central oncogenesis pathway genesis in tumors have uncovered a bewilderingly large or in parallel pathways that provide the cell with equiva- number of genes that bear the signature of genetic selec- lent functions to drive tumorigenesis. Thus, it is likely tion in tumors. Rather than a defined number of cancer that many genes on these TSG and OG lists will genetical- drivers, there exists a continuum of genes that appear ly interact to modify common conditions of oncogenesis. with increasingly lower frequency and potency in a pan- One of the most extensively studied oncogenic path- cancer analysis (Davoli et al. 2013). This has also been re- ways is the receptor tyrosine kinase (RTK)–RAS–phos- ferred to as mountains and hills and the “long tail” of can- phoinositol-3-kinase (PI3K) pathway. This pathway is cer drivers (Wood et al. 2007; Leiserson et al. 2015; Cho activated in the majority of solid tumors and has been et al. 2016). examined extensively both biochemically and genetical- Unraveling how the genes in this large network contrib- ly. Among RTKs, perhaps the most studied is the epider- ute to tumorigenesis poses a significant challenge for mal growth factor receptor (EGFR). EGFR activates geneticists. While it is clear that certain basic conditions cellular signaling pathways such as PI3K/AKT, RAS/ related to the classic hallmarks of cancer (e.g., immortal- RAF/MEK/ERK, and JAK/STAT, leading to increased ization and deregulated proliferation) must be satisfied during tumor development (Hanahan and Weinberg © 2017 Liao et al. This article is distributed exclusively by Cold Spring Harbor Laboratory Press for the first six months after the full-issue publi- cation date (see http://genesdev.cshlp.org/site/misc/terms.xhtml). After Corresponding author: [email protected] six months, it is available under a Creative Commons License (At- Article published online ahead of print. Article and publication date are tribution-NonCommercial 4.0 International), as described at http://creati- online at http://www.genesdev.org/cgi/doi/10.1101/gad.291948.116. vecommons.org/licenses/by-nc/4.0/. GENES & DEVELOPMENT 31:1–13 Published by Cold Spring Harbor Laboratory Press; ISSN 0890-9369/17; www.genesdev.org 1 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press Liao et al. cell proliferation and survival (Chong and Janne 2013). Ac- NSCLC cells in part through sustained activation of tivating EGFR mutations occur in ∼10%–30% of tumors ETV1, resulting in gefitinib resistance. These findings pro- of patients with non-small cell lung cancer (NSCLC), a vide new biochemical insight into EGFR signaling and leading cause of cancer-related deaths (Stewart et al. support the general notion that cancer drivers are part of 2015). These mutations confer sensitivity to EGFR inhib- a robust joint network that can compensate for the loss itors (EGFRis) such as gefitinib and a variety of later-gen- of any one member. eration inhibitors (Lynch et al. 2004; Paez et al. 2004; Wang et al. 2016). Although EGFR mutant NSCLCs typi- cally respond dramatically to EGFRis, these responses are Results not universal, as the overall response rate is ∼71%. Even among the initial responders, most inevitably develop ac- The central hypothesis motivating this study is that can- quired resistance to EGFRi therapies within a year of treat- cer mutations often impact the same pathways or control ment (Mok et al. 2009; Rosell et al. 2009; Thress et al. parallel pathways that can substitute for each other. To 2015). The resistance mechanism is unknown in up to test this notion, we investigated the RTK EGFR pathway. 30% of patients (Majem and Remon 2013). While EGFR has been extensively studied both biochemi- Given its central role in driving oncogenesis, the exist- cally and genetically, it has not been systematically ing knowledge of the pathway, and the many tools avail- probed for its interactions with all known and putative able, the EGFR pathway is well suited for examining cancer drivers. To explore these interactions for genes genetic interactions with other known and putative with TSG properties, we generated both a CRISPR and cancer drivers. This is supported by existing evidence of an shRNA library containing 10 guide RNAs (gRNAs) or genetic interactions of EGFR with other drivers of tumor- 10 shRNAs per gene to a list of ∼500 genes whose LOF igenesis (Sharifnia et al. 2014). For example, patients bear- has been implicated in driving tumorigenesis by the ing EGFR mutations are known to evolve resistance to TUSON Explorer algorithm (Davoli et al. 2013). Each li- EGFRi therapies by virtue of mutations in other cancer brary also contained 1000 gRNAs or 1000 shRNAs target- drivers. In addition to mutations in EGFR itself, low ex- ing the Escherichia coli genome as negative controls. To pression of NF1 (de Bruin et al. 2014) or PTEN (Sos et al. explore the genetic interactions with EGFR for genes 2009; Yamamoto et al. 2010), amplification of the MET with OG properties, we generated a barcoded ORF lentivi- RTK (Engelman et al. 2007), amplification of the HER2 rus library of ∼50 selected genes whose mutational signa- (ERBB2) RTK (Takezawa et al. 2012), and activation of tures implicate them as potential OGs by TUSON KRAS, PIK3CA, BRAF, and MAPK1 (ERK) (Sartore-Bian- Explorer (Fig. 1A; Davoli et al. 2013). We set out to deter- chi et al. 2009; Diaz et al. 2012; Ercan et al. 2012; Misale mine which alterations could substitute for EGFR signal- et al. 2012; Ohashi et al. 2012) can confer EGFRi resis- ing using a chemical inhibitor of EGFR, gefitinib. We tance. Thus, it is likely that additional drivers will also ge- performed screens using a NSCLC cell line, PC9, which netically interact with the EGFR pathway. harbors an activating EGFR mutation and is sensitive to To test the hypothesis that cancer drivers can genetical- gefitinib. CRISPR and shRNA have different mechanisms ly interact and substitute for one another to drive prolifer- and off targets, thus providing complementary means of ation and survival, we investigated TSG and OG drivers assessing the functional contribution of TSGs to EGFRi for their ability when mutated to partially replace EGFR resistance. Genes that retain function at low expression in EGFR-dependent tumor cells by performing CRISPR, levels are likely to be missed in shRNA screens due to shRNA, and OG expression screens in parallel in a their incomplete depletion. In contrast, genes that are es- NSCLC model. We took advantage of an algorithm called sential for cell viability cannot be assessed in CRISPR TUSON (Tumor Suppressor and Oncogene) Explorer to screens. Partial depletion by shRNA will be useful in identify TSGs and OGs (Davoli et al. 2013). This method these cases. In addition, as gene regulatory networks are quantifies the likelihood that a gene is a cancer driver highly interconnected and contain multiple feedback based on the distortion of its mutational signature from loops, the response to knockout and depletion can be the pattern expected for a “neutral” gene. For example, markedly different (Shalem et al. 2015). By performing TSGs will have higher ratio of loss of function (LOF) to these complementary CRISPR and shRNA screens in par- benign mutations than neutral genes (Fig.